Is there a where like relation function when using pymongo?

Question

I have the following data:

A collection named categories that contains documents like so:

{
    "cat":"Films Anglais",
    "path":"W:\\videos"
}

Categories are unique (because I use upsert) or let's admit it is anyway.

A collection named rules that contains documents like so:

{
    "title":"braveheart",
    "regex":"^.*braveheart.*$",
    "cat":"Films Anglais"
}

I am iterating on all the rules. So I can access the cat from rules as rule['cat']. What I need is the path from categories.

I know that I can do:

dest = ""
for category in categories.find():
    if category['cat'] == rule['cat']:
        dest = category['path']
        break

1) I would prefer the process to be database side. Like categories.find_one().distinct('path').where(cat=rule['cat'])? Invalide I know.

2) Is there a way to define a sort of relation so that I do not need to duplicate the cat field?

Lastly, I have read about the difference between relationnal and non relationnal systems but in this case the choice is sealed.

Answer 1

For 1), you don't want to use server-side javascript here, or ever, really. It's slow and blocks lots of other operations. Don't use server-side javascript to try and fake joins in MongoDB.

For 2), duplicating the path info into the rules documents seems like the best solution. How often will a path change? The cost to embedding path is the duplication and the need for a more expensive set of updates if a path changes, as compared to your current setup. Seems worth it to me, in the absence of further information about your use case.